Frequently asked questions
What is text to speech in transit?
Text-to-speech converts written transit announcements into audio files via a simple API. You send text and receive synthesized speech suitable for station displays, app notifications, or digital signage. It replaces manual recording with automated, consistent audio generation that integrates directly into your transport product workflows.
How does the API work?
Send a POST request to the /v1/audio/speech endpoint with your text, voice, and format. The API returns an audio file. You can use the official OpenAI SDK by changing the base URL to https://api.playc54.com/v1. Authentication uses a Bearer token from your account dashboard.
Which voice is best for announcements?
Atlas offers a neutral, news-style tone ideal for clear transit updates. Sage provides a calm, instructional voice suitable for detailed instructions. Aria is a warm narrator, while Felix is energetic and suitable for promos. Choose based on whether you need neutral information or engaging announcements.
Can I use the API for live updates?
The API generates audio files synchronously; it does not support real-time streaming or WebSocket connections. Generation takes a few seconds for short phrases. For live updates, generate the audio file ahead of time or trigger generation slightly before the announcement is needed.
How do I handle multiple languages?
Specify the ISO language code in the request body. If omitted, the system guesses the language, defaulting to English for Latin script. We support 23 languages, including English, Spanish, German, and Japanese. Ensure you set the language parameter for accurate pronunciation in non-English texts.
Is the audio quality high enough for stations?
The audio is clear and natural, suitable for digital signage and apps. Files include an inaudible watermark marking them as synthetic. Use the tts-1-hd model for higher fidelity. The quality is consistent across all 8 voice presets, ensuring professional-sounding transit announcements.
Can I control the speaking speed?
Yes, adjust the speed between 0.25 and 4.0. This allows you to slow down announcements for clarity or speed up brief notifications. The default speed is 1.0. Adjusting speed does not affect the voice character or the underlying synthesis engine quality.
Do I need to format my text?
No special formatting like SSML is required. Just send plain text in the input field. The API handles punctuation and pauses naturally. You can include standard punctuation to influence intonation. Keep text under 4096 characters per request for optimal processing.
How long does generation take?
Generation is roughly real-time, averaging 25 characters per second. Short phrases take a few seconds. Longer texts are processed sequentially, so high load may cause queueing. Failed requests are free and do not incur charges or delays to your balance.
Can I download the audio as MP3?
Yes, select mp3 as the response_format. You can also choose wav, opus, flac, or pcm. The API returns the audio file directly in the response. You can then save or play the file in your transport application or digital signage system.
Is there a limit on characters per request?
Each request supports up to 4096 characters. For longer announcements, split the text into parts, generate audio for each, and join the files. Spaces and punctuation count toward the character limit. This ensures efficient processing and consistent audio quality for your transit messages.
How do I know which language is used?
Set the language parameter to the ISO code for precise control. If unset, the system guesses from the script, defaulting to English for Latin text. Explicitly setting the language ensures correct pronunciation, especially for non-English languages using Latin scripts like Turkish.
Can I use my own text data?
Yes, send any text up to 4096 characters. Your input is used only to produce the audio and is not used for training. Generated audio is returned in the response and not stored. You retain full control over the content and timing of announcements.
Is the API secure for production?
Use a Bearer token for authentication. Each account has one active key, which can be revoked instantly. We do not store generated audio or use input text for training. Request metadata is kept for billing. There is no SLA or uptime guarantee for production use.
Understanding TTS for Transit
Text-to-speech technology has transformed how transit systems communicate with passengers. By converting digital text into natural-sounding audio, operators can deliver timely updates without recording every variation manually. This ensures consistency across station announcements, app notifications, and digital signage displays.
The system handles text up to 4096 characters per request, allowing for comprehensive instructions in a single file. With speeds adjustable from 0.25x to 4.0x, announcements can be tailored to match the urgency of the situation, whether it is a calm station ID or a quick gate change.
Clear audio standards
Our API delivers high-quality audio in multiple formats, including MP3, WAV, Opus, FLAC, and PCM. This flexibility ensures compatibility with various playback systems, from legacy station speakers to modern mobile apps. Every generated file includes an inaudible watermark to identify synthetic speech.
Using voices like Atlas for news or Sage for instructional content provides a professional tone suitable for public transport environments.
- Support for MP3, WAV, Opus, FLAC, and PCM formats
- Inaudible watermark marks files as synthetic speech
- Adjustable speed from 0.25x to 4.0x for pacing control
Noise resistance
Transit environments are often noisy, with background sounds from trains, crowds, and announcements overlapping. ClearCast’s voices are designed for clarity, but intelligibility in challenging acoustic conditions depends on your playback hardware. The natural cadence and clear enunciation help passengers catch essential details despite ambient noise.
For optimal results, use a calm voice like Luna for storytelling or Milo for conversational updates. These voices maintain clarity without sounding overly robotic, helping passengers distinguish announcements from background chatter. Testing audio in actual environments is recommended to verify readability.
- Voices optimized for clarity in noisy settings
- Natural cadence helps cut through background sounds
- Adjustable speed allows for slower, clearer delivery in loud areas
User experience
Passengers appreciate consistent and timely information. By automating announcements, transit operators can ensure that updates are delivered instantly, reducing anxiety during delays or route changes. The ability to generate audio on demand means last-minute schedule changes can be communicated immediately.
A positive user experience relies on reliability. Our API processes requests sequentially, ensuring accurate delivery. With a free trial of $0.50, product teams can test different voices and languages to find the best fit for their specific audience.
- Instant generation for last-minute schedule changes
- Consistent voice branding across all stations
- Free trial allows testing before full deployment
Choosing the Right Voice
Selecting the right voice ensures your transit announcements are both clear and appropriate for the context. Whether you need a neutral broadcast style or a warmer tone for customer updates, matching the voice to the message improves user experience.
Our API provides eight distinct presets, each optimized for specific use cases. You can switch voices instantly via the voice parameter in your API request, allowing you to tailor announcements for different stations, times of day, or audience segments.
Neutral news style
For station updates and emergency alerts, clarity is paramount. Atlas provides a neutral, news-ready tone that cuts through background noise effectively. It is ideal for time-sensitive information where the message matters more than personality.
Use Atlas for standard train arrival times, platform changes, or safety instructions. Its consistent delivery ensures listeners grasp critical details quickly, even in busy terminal environments.
- Ideal for: real-time updates, safety warnings, platform changes
Friendly conversational
For app notifications or loyalty program updates, a warmer tone builds connection. Milo offers a friendly, conversational style that feels less like a broadcast and more like a helpful assistant.
This approach works well for personalized messages, such as greeting returning passengers or announcing special promotions. It softens the technical nature of automated systems, making digital signage feel more welcoming.
- Ideal for: app notifications, promotional messages, welcome greetings
Calm instructional
When guiding users through complex processes, Sage’s calm, instructional tone reduces cognitive load. It is particularly effective for step-by-step instructions, such as navigating a new ticketing system or boarding procedures.
Use Sage when precision and patience are required. Its steady pace helps listeners follow along without feeling rushed, making it a reliable choice for detailed operational guidance.
- Ideal for: boarding instructions, navigation guides, step-by-step tutorials
Setting Up Your API Key
Access to the ClearCast API requires an active account tied to the specific site you are using. You can create an account by visiting the /get-api-key/ page, where you can sign in using Google or register with an email and password. This step establishes your unique identity for billing and usage tracking.
Once registered, you are issued a single active API key starting with 'tts-'. This key is your only credential for authentication. You can generate a new key at any time, which immediately invalidates the previous one, ensuring you always control access to your account.
Sign up on site
Navigate to the /get-api-key/ page on playc54.com. Choose to sign in with Google for quick access or register using an email address and password. Accounts are per site, so ensure you are on the correct domain.
After registration, your account is created. If you used Google, you receive a free trial credit of $0.50 valid for 7 days. Email-based accounts start without trial credit. You are now ready to retrieve your credentials.
- Visit https://playc54.com/get-api-key/
- Choose Google sign-in or email registration
- Receive account confirmation instantly
Copy your key
After signing up, the dashboard displays your API key. It begins with the prefix 'tts-'. Copy this entire string carefully, as it is required for all API requests. The key is shown on the dashboard alongside your current balance and usage statistics.
Note that each account has only one active key at a time. If you lose your key or suspect it is compromised, you can issue a new one from the dashboard. The old key stops working immediately upon generation of the new one.
- Locate the key on the dashboard
- Copy the full string starting with 'tts-'
- Store it securely in your environment variables
Secure storage
Treat your API key like a password. Store it in environment variables or a secure secrets manager rather than hardcoding it in your application source code. This prevents accidental exposure if your code repository becomes public. Use the header 'Authorization: Bearer <your-key>' for requests.
Since there is only one key per account, ensure your development, staging, and production environments share or securely manage this credential appropriately. Regenerate the key if you suspect unauthorized usage, which immediately revokes the previous key.
- Use environment variables for storage
- Include in 'Authorization: Bearer' header
- Regenerate if compromised
Making Your First Request
To generate transit audio, send a POST request to https://api.playc54.com/v1/audio/speech. The API accepts JSON, allowing you to define the model, text, voice, and format in a single call. This structure mirrors standard TTS endpoints, making integration straightforward for your transport product team.
Construct JSON body
Your request body must include the model ("tts-1" or "tts-1-hd"), the input text (up to 4096 characters), and the voice preset. You can also specify response_format (mp3, wav, etc.) and speed (0.25–4.0). Include an Authorization: Bearer header with your API key.
- model: "tts-1" or "tts-1-hd"
- input: your announcement text
- voice: e.g., "atlas" for news
- response_format: "mp3" or "wav"
Set parameters
Choose a voice that fits the context: "atlas" for neutral station announcements or "milo" for friendly conversational updates. Set the speed between 0.25 and 4.0 to adjust playback rate. If the text is in a specific language, provide the ISO code to ensure correct pronunciation, though the system guesses from the script otherwise.
- Use "atlas" for clear, neutral news
- Set speed between 0.25 and 4.0
- Specify
languagefor non-English text
Send POST request
Send the request to https://api.playc54.com/v1/audio/speech. The API returns the audio file directly in the response. Failed requests are free, ensuring you only pay for successful generations. You can then save the file or play it in your app.
Handling the Response
Once the API returns a success status, the response body contains the raw audio data. The format matches your requested response_format, typically MP3 for compatibility. This binary data is the complete audio file ready for use.
Download audio file
The API returns the audio file directly in the response body. No separate download link is generated; the data is immediate and ready to be written to a file or played by a player.
- Supports MP3, WAV, OPUS, FLAC, and PCM formats
- No additional download steps required
- Immediate availability upon successful synthesis
Save to storage
Store the returned audio file in your preferred cloud storage or local server. Since the API does not persist generated audio, you are responsible for saving files if you need long-term access or replayability.
- Files are not stored by the API
- Recommended for caching repeated announcements
- Save as MP3 for optimal size
Play in app
Integrate the audio file into your station announcement system or mobile app. Use standard HTML5 audio tags or native media players to play the file. The audio is ready for immediate playback in digital signage or app interfaces.
- Compatible with standard media players
- Ideal for digital signage displays
- Works in mobile and web apps
Optimizing for Speed
Speed optimization ensures transit announcements play without awkward pauses or rushed delivery. By adjusting the speed parameter between 0.25 and 4.0, you can fine-tune audio to match ambient station noise or app notification contexts. Faster speeds suit brief alerts, while slower rates improve clarity for complex route information.
Pre-generating content during off-peak hours and caching results reduces latency during high-traffic moments. Since requests are processed sequentially, batching shorter announcements helps maintain steady throughput. Caching allows your system to serve pre-rendered audio files instantly, bypassing API queues entirely when the same announcement repeats frequently.
Adjust speed parameter
Set speed values between 0.25 and 4.0 to control playback rate. Higher speeds compress audio duration, useful for quick app notifications. Lower speeds enhance intelligibility for detailed station instructions.
The API processes requests sequentially, so higher speeds may reduce total generation time but do not bypass queue limits. Test different speeds with your specific text to find the optimal balance between clarity and duration.
- Use 1.0 for standard narration
- Increase to 1.5–2.0 for quick alerts
- Decrease to 0.75–0.9 for complex route details
Pre-generate content
Generate announcements during low-traffic periods to avoid queue delays. Split long texts into segments under 4096 characters before processing to ensure consistent generation times.
Combine pre-generated audio files to create custom sequences. This method avoids real-time synthesis overhead for scheduled broadcasts or static digital signage displays.
- Batch requests during off-peak hours
- Split texts under 4096 characters
- Combine files for scheduled broadcasts
Cache results
Store generated MP3 files locally or in your cloud storage. Serve cached files directly to users instead of calling the API for identical announcements. This eliminates generation time and API costs for repeated content.
Implement cache invalidation strategies to update announcements when route changes occur. Use unique filenames or versioning to manage updates without disrupting existing user experiences.
- Store MP3s in cloud storage
- Serve cached files directly to users
- Use versioning for updates

Testing Different Voices
Choosing the right voice is critical for passenger trust. ClearCast provides eight presets designed for specific tones, allowing product teams to match the announcement style to the context. Each voice is based on licensed recordings, ensuring consistent quality across all generated audio files.
You can test these voices directly via the API to hear how they sound in your app’s environment. This helps you select the best fit for your users' needs.
Aria for narrations
Aria offers a warm, clear narrator voice ideal for longer transit updates or station stories. Her tone is engaging without being distracting, making her suitable for detailed route information or service changes.
Use Aria when you need to hold the passenger’s attention for more than just a simple alert.
- Warm, clear narrator
- Ideal for detailed updates
- Engaging tone
Atlas for news
Atlas delivers a neutral, news-style voice perfect for urgent or factual announcements. His sound is professional and authoritative, ensuring that important information like delays or platform changes is communicated clearly.
This voice works best for time-sensitive alerts where clarity and neutrality are prioritized over personality.
- Neutral, news style
- Authoritative tone
- Best for urgent alerts
Milo for chats
Milo provides a friendly, conversational voice suitable for interactive features or less formal updates. His approachable tone can make routine announcements feel more personal and less robotic.
Use Milo for app notifications or help-desk style interactions where a lighter tone is appropriate.
- Friendly, conversational
- Approachable tone
- Good for notifications
Managing Languages
Transit environments often serve diverse passenger groups, making precise language control essential for clarity. While the API can guess the language from the script, explicitly setting the language parameter ensures reliable pronunciation, especially for non-English texts using Latin characters.
Set language code
Always provide the ISO language code (e.g., fr, de, es) in your request. This prevents misinterpretation and ensures the voice renders accents correctly, which is critical for wayfinding and safety announcements.
- Explicit codes reduce ambiguity for similar scripts
- Ensures correct accent and intonation
- Required for non-English Latin text
Auto-detect script
If you omit the language field, the system guesses based on the script. Latin script defaults to English, while other scripts trigger their respective models. This works well for single-language content but may fail for mixed-language announcements.
Multi-language support
Our voices support 23 languages, allowing you to serve diverse routes with consistent audio quality. Each voice can speak any supported language, ensuring uniform tone across different stations or lines without needing multiple voice assets.
- 23 supported languages including Arabic, Chinese, Hindi, and Japanese
- No need for separate voice models per language
- Consistent tone across diverse passenger demographics
Monitoring Usage
Track your API usage to stay within budget and avoid service interruptions. The dashboard provides a clear view of your current balance, characters consumed, and request history, ensuring you always know your limits.
Regular monitoring helps you optimize text length and speed settings for efficient generation, keeping your transit announcements running smoothly without unexpected charges.
Check balance
View your current account balance directly on the dashboard. Your balance does not expire, and you can top up with USDT on the TRON network at any time. A new key invalidates the old one instantly, so keep your active key secure.
- Real-time balance updates
- Instant top-up via TRC-20
- No expiration on funds
Track characters
Each request counts characters toward your limit, including spaces and punctuation. You start with 10 requests per minute, scaling to 60 requests per minute after your first top-up. Failed requests are free, so only successful generations deduct from your balance.
- Spaces and punctuation count
- Failed requests are free
- 60 requests/minute limit
View history
Review your request history to see timestamps, character counts, and status for each generation. This data helps you audit usage and troubleshoot any issues with specific announcements or voices.
- Request timestamps
- Character count per request
- Success/failure status
Troubleshooting Common Issues
When your transit announcements fail to generate, check your text length first. Each request supports up to 4,096 characters. If your announcement is longer, split it into parts and join the audio files. Failed requests are free, so you won’t lose credit if the synthesis fails.
If you hit the rate limit of 60 requests per minute, wait a moment before retrying. New accounts start with 10 requests per minute. Ensure your API key is active and correctly formatted in the authorization header.
Audio generation fails
Check that your input text does not exceed 4,096 characters per request. Split longer announcements into smaller chunks and concatenate the resulting audio files. Ensure your API key is valid and active.
- Limit text to 4,096 characters per request
- Split long announcements and join audio
- Verify active API key
Rate limit reached
New accounts can make 10 requests per minute. Once you top up, the limit increases to 60 requests per minute. Monitor your usage in the dashboard to avoid interruptions during peak announcement times.
- 10 req/min for new accounts
- 60 req/min after top-up
- Monitor usage in dashboard
Incorrect voice
Ensure you are selecting a voice that matches your desired tone. For neutral news, use Atlas. For friendly conversations, use Milo. Verify the voice ID in your request matches the available presets.
- Use Atlas for neutral news
- Use Milo for friendly chats
- Verify voice ID in request
Best Practices for Announcements
Clear transit announcements require precise input to avoid audio artifacts that distort in noisy stations. By keeping text under the 4,096-character limit and using standard punctuation, you ensure the TTS engine generates consistent, crisp audio files ready for digital signage or app notifications.
Keep text concise
Transit passengers often listen while moving or in crowded areas. Short, direct sentences reduce cognitive load and ensure the message is understood instantly. Avoid complex clauses that might confuse the synthesis engine or dilute the clarity of the announcement.
- Use short sentences for quick reading.
- Limit text to essential information.
- Avoid complex grammatical structures.
Use clear punctuation
Punctuation directly controls the pacing and intonation of the synthetic voice. Proper commas and periods create natural pauses, preventing the audio from sounding rushed or monotone. This is crucial for maintaining a professional tone in station announcements.
- Use periods for full stops.
- Use commas for natural pauses.
- Avoid excessive exclamation marks.
Test in noisy environments
Generate audio samples and play them in the actual deployment environment, such as a busy platform or a moving vehicle. This helps verify that the volume and clarity are sufficient for passengers to hear the announcement clearly over ambient noise.
- Play audio at maximum volume.
- Test during peak hours.
- Check background noise levels.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Aria delivers a soft, storytelling quality that works well for detailed route explanations or station history segments.
23 languages
Any voice speaks any of these languages.
